求mnist多数字识别，修改完成我的代码
实现多个 数字的识别 如 
实现方法： 把五个数字的图 拼成一个 再进行 训练 和 测试
学校讲的自己看的都一知半解，我都不知道 我一个大二的，没什么基础的学生是怎么选上做这个研究的。。马上due就快到了，求大神在我代码基础上帮我完成。。
import tensorflow as tf
import numpy as np
from numpy import array
import os
import cv2
import matplotlib.pyplot as plt
mnist = tf.keras.datasets.mnist
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train = tf.keras.utils.normalize(x_train, axis=1)
x_test = tf.keras.utils.normalize(x_test, axis=1)
import random
countain = 0
myxtrain = []
myytrain = []
while countain < 5:
a_1 = random.randint(0, 10000)
a_2 = random.randint(0, 10000)
a_3 = random.randint(0, 10000)
a_4 = random.randint(0, 10000)
a_5 = random.randint(0, 10000)
a = np.concatenate((x_train[a_1], x_train[a_2], x_train[a_3], x_train[a_4], x_train[a_5]), axis=1)
myxtrain.append(a)
labelx = []
s1 = str(y_train[a_1])
s2 = str(y_train[a_2])
s3 = str(y_train[a_3])
s4 = str(y_train[a_4])
s5 = str(y_train[a_5])
labelx.append(s1)
labelx.append(s2)
labelx.append(s3)
labelx.append(s4)
labelx.append(s5)
s = ' '.join(labelx)
print('----', s)
# cv2.imwrite(os.path.join(path1, s + '.jpg'), a)
# cv2.waitKey(0)
myytrain.append(s)
countain += 1
x_train = array(myxtrain)
y_train = array(myytrain)
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D
from keras.optimizers import SGD
model = tf.keras.models.Sequential()
model.add(Conv2D(5, (3, 3), activation='relu', input_shape=(28, 140, 2)))
model.add(Conv2D(5, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten())
model.add(Dense(256, activation='relu'))
model.add(Dense(10, activation='softmax'))
sgd = SGD(lr=0.0001, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='categorical_crossentropy', optimizer=sgd)
model.fit(x_train, y_train, batch_size=1, epochs=10)
关键代码如下：
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("mnist_set",one_hot=True)
#数据集相关常数
INPUT_NODE = 784
OUTPU_NODE = 10
#配置神经网络参数
LAYER1_NODE = 500
BATCH_SIZE = 100
LEARNING_RATE_BASE = 0.8 #基础学习率
LEARNING_RATE_DECAY = 0.99#学习衰减率
REGULARIZATION_RATE = 0.0001#正则的惩罚系数
MOVE_AVG_RATE = 0.99 #滑动平均衰减率
TRAIN_STEPS = 30000
def inference(input_tensor,weights1,biases1,weight2,biases2,avg_class=None):
#当没有提供滑动平均类时，直接使用当前值
if avg_class == None:
#计算隐藏层的前向传播结果，使用RELU激活函数
layer1 = tf.nn.relu(tf.matmul(input_tensor,weights1)+biases1)
#返回输出层的前向传播
return tf.matmul(layer1,weight2)+ biases2
else:
#前向传播之前，用avg——class计算出变量的滑动平均值
layer1 = tf.nn.relu(tf.matmul(input_tensor,avg_class.average(weights1))+avg_class.average(biases1))
return tf.matmul(layer1,avg_class.average(weight2))+ avg_class.average(biases2)
#模型的训练过程
def train(mnist):
x = tf.placeholder(tf.float32,[None,INPUT_NODE],name='x-input')
y_ = tf.placeholder(tf.float32,[None,OUTPU_NODE],name="y_input")
#隐藏层参数
w1 = tf.Variable(tf.random_normal([INPUT_NODE,LAYER1_NODE],stddev=0.1))
b1 = tf.Variable(tf.constant(0.1,shape=[LAYER1_NODE]))
#输出层参数
w2 = tf.Variable(tf.random_normal([LAYER1_NODE,OUTPU_NODE],stddev=0.1))
b2 = tf.Variable(tf.constant(0.1,shape=[OUTPU_NODE]))
y = inference(x,w1,b1,w2,b2)
#定义存储训练轮数的变量，设为不可训练
global_step = tf.Variable(0,trainable=False)
#初始化滑动平均类
variable_averages = tf.train.ExponentialMovingAverage(MOVE_AVG_RATE,global_step)
#在神经网络的所有参数变量上使用滑动平均
variable_averages_op = variable_averages.apply(tf.trainable_variables())
averages_y = inference(x,w1,b1,w2,b2,avg_class=variable_averages)
#计算交叉熵损失
cross_entropy = tf.nn.softmax_cross_entropy_with_logits(logits=y,labels=tf.argmax(y_,1))
#计算交叉熵平均值
cross_entropy_mean = tf.reduce_mean(cross_entropy)
#计算L2正则的损失函数
regularizer = tf.contrib.layers.l2_regularizer(REGULARIZATION_RATE)
regularization = regularizer(w1)+ regularizer(w2)
#总的损失
loss = cross_entropy_mean + regularization
#设置指数衰减的学习率
training_rate = tf.train.exponential_decay(LEARNING_RATE_BASE,global_step,mnist.train.num_examples/BATCH_SIZE,LEARNING_RATE_DECAY)
train_step = tf.train.GradientDescentOptimizer(training_rate).minimize(loss,global_step)
#更新滑动平均值
with tf.control_dependencies([train_step,variable_averages_op]):
train_op = tf.no_op(name='train')
#验证前向传播结果是否正确
correct_prediction = tf.equal(tf.argmax(averages_y,1),tf.argmax(y_,1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
#初始会话，开始训练
with tf.Session() as sess:
tf.global_variables_initializer().run()
#准备验证数据和测试数据
validate_feed = {x:mnist.validation.images,y_: mnist.validation.labels}
test_feed = {x:mnist.test.images,y_: mnist.test.labels}
#迭代训练神经网络
for i in range(TRAIN_STEPS):
if i % 1000 == 0:
validate_acc = sess.run(accuracy,feed_dict=validate_feed)
print("After %s training steps ,validation accuracy is %s"%(i,validate_acc))
xs,ys = mnist.train.next_batch(BATCH_SIZE)
sess.run(train_op,feed_dict={x:xs,y_:ys})
#训练结束后，在测试集上验证准确率
test_acc = sess.run(accuracy,test_feed)
print("After %s training steps ,test accuracy is %s"%(TRAIN_STEPS,test_acc))
def main(argv=None):
mnist = input_data.read_data_sets("mnist_set",one_hot=True)
train(mnist)
if __name__ == '__main__':
#TF 提供了一个主程序入口，tf.app.run会自动调用上面的main（）
tf.app.run()
完整代码和训练数据模型要等采纳以后给你。